Harvard Law Review Harvard Law Review Harvard Law Review

State Action Doctrine

In re Late Fee and Over-Limit Fee Litigation

Ninth Circuit Rejects Due Process Limit on Credit Card Fees.

In recent decades, the Supreme Court has held that excessive punitive damage awards in tort suits violate the Due Process Clause.1×1. See State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408 (2003); BMW of N. Am., Inc. v. Gore, 517 U.S. 559 (1996). Consumer advocates have invoked the Court’s punitive damages decisions to challenge contractual credit card penalty fees as similarly violative of due process.2×2. See generally, e.g., Seana Valentine Shiffrin, Are Credit Card Late Fees Unconstitutional?, 15 Wm. & Mary Bill Rts. J. 457 (2006) (proposing and defending this legal theory). Recently, in In re Late Fee and Over-Limit Fee Litigation,3×3. 741 F.3d 1022 (9th Cir.), cert. denied sub nom. Piñon v. Bank of Am., NA, 134 S. Ct. 2878 (2014). the Ninth Circuit rejected such a claim as unsupported by precedent, but a majority of the panel encouraged the Supreme Court to extend the tort cases’ reasoning to contract cases involving credit card agreements.4×4. See id. at 1028 (Reinhardt, J., concurring in the judgment). The state action requirement, however, poses a principal difficulty to this extension. Revitalizing the public function theory of state action,5×5. The public function theory finds state action when private entities perform functions that have traditionally been exclusively governmental. See Erwin Chemerinsky, Constitutional Law § (4th ed. 2011). despite its current dormancy, would be one means of protecting consumers in an age in which many traditional contract law protections have been supplanted by boilerplate terms, narrowed by courts, or abrogated by positive law. Public function theory, while absent from both the briefing and opinions, is the only theory of state action that captures an important feature of why punitive credit card fees are objectionable: they entail card issuers exercising state-like power over consumers.

Provisions allowing card issuers to impose penalty fees, including fees for late payments or over-limit use, are included in the contracts necessary for obtaining credit cards.6×6. See Shiffrin, supra note 2, at 481. Although recent federal legislation has resulted in limits on the amount of some credit card fees,7×7. The Credit CARD Act of 2009, Pub. L. No. 111-24, 123 Stat. 1734 (codified as amended in scattered sections of 15, 16, and 31 U.S.C.), requires that fees be “be reasonable and proportional,” 15 U.S.C. § 1665d(a) (2012). See also 12 C.F.R. 1026.52(b)(ii) (2014) (limiting late fees to $26 for a first late payment and $37 for subsequent late payments). fees still constitute more than one-third of issuers’ annual revenue.8×8. See Daniel Carusotto, IBISWorld Industry Report 52221, Credit Card Issuing in the US 4 (2014). A slightly late payment or slightly over-limit use imposes only de minimis marginal costs on issuers,9×9. See Shiffrin, supra note 2, at 478–79. and card issuers recoup any costs through raising cardholders’ interest rates.10×10. See id. at 477–78. As a result, the fees are best understood not as compensatory, but rather as either driven by profit motives or as “purely punitive.”11×11. See id. at 479. Credit card fees in excess of a 100:1 ratio of compensatory to punitive damages, see id., are in tension with State Farm’s pronouncement that “few [tort] awards exceeding a single-digit ratio between punitive and compensatory damages” will satisfy due process. State Farm Mut. Auto Ins. Co. v. Campbell, 538 U.S. 408, 425 (2003). Such fees are forbidden under the common law of contracts,12×12. See Restatement (Second) of Contracts § 356(1) (1981) (“Damages for breach by either party may be liquidated . . . only at an amount that is reasonable in the light of the anticipated or actual loss caused by the breach and the difficulties of proof of loss. A term fixing unreasonably large liquidated damages is unenforceable on grounds of public policy as a penalty.”). which has traditionally been particularly suspicious of late fees on the grounds that they are often punitive rather than compensatory.13×13. See 11 Joseph M. Perillo, Corbin on Contracts § 58.13 (rev. ed. 2005). Credit card penalty fees have been permitted, however, by positive law abrogating the common law of contracts.14×14. Through a combination of state and federal statutes, credit card fees are not subject to the presumption against punitive contract damages. See Shiffrin, supra note 2, at 466–67; see also 12 U.S.C. § 85 (2012) (permitting banks to charge interest); 12 C.F.R. § 7.4001(a) (2014) (including penalty fees under the statute’s definition of interest); Smiley v. Citibank (S.D.), N.A., 517 U.S. 735 (1996) (allowing banks to charge customers late fees at rates permitted by banks’ home states, even if in excess of rates permitted by the laws of the customers’ home states).

In January 2007, cardholders sued card issuers in the Northern District of California, alleging, inter alia, that contractual penalty fees violate due process limits on punitive damages.15×15. See Complaint at 1–3, In re Late Fee & Over-Limit Fee Litig., 528 F. Supp. 2d 953 (N.D. Cal. 2007) (No. 4:07-cv-00634). The complaint, filed by a class of cardholders who had paid late fees and/or over-limit fees, see id. at 1, also included claims under federal banking law, federal and state antitrust law, and state contract law, see id. at 26–35. In an opinion by Judge Armstrong, the district court dismissed the complaint,16×16. See Late Fee & Over-Limit Fee Litig., 528 F. Supp. 2d at 953. rejecting all of the plaintiffs’ claims, including their attempt to apply the Due Process Clause to credit card penalty fees.17×17. Id. at 959. It reasoned that credit card fees are not subject to the Due Process Clause because they are merely “paid by one party to another pursuant to private contract,”18×18. Id. do not advance “governmental objectives,”19×19. Id. (quoting Pac. Mut. Life Ins. Co. v. Haslip, 499 U.S. 1, 47 (1991) (O’Connor, J., dissenting)) (internal quotation mark omitted). and do not impose the risks, present in the jury awards context, of “arbitrariness, uncertainty, and lack of notice.”20×20. Id. (quoting Philip Morris USA v. Williams, 549 U.S. 346, 354 (2007)) (internal quotation mark omitted). Additionally, it found that credit card fees — “even if [they] could be regarded as some kind of private ‘punitive damages’” — do not implicate state action and therefore do not violate the Due Process Clause.21×21. Id. at 960. The Court distinguished fees set by government mandate from those resulting from private contract, and rejected the plaintiffs’ argument that contractual penalty fees are state action when affirmatively permitted by a federal regulatory framework.22×22. See id. The court also rejected the plaintiffs’ claim that constitutional avoidance compels construing the National Bank Act’s usury provision, 12 U.S.C. § 85 (2012), as incorporating due process limits. See id. at 960–61. In doing so, it cited the lack of ambiguity of both the statutory text and the relevant constitutional issue. See id.

The Ninth Circuit affirmed.23×23. Late Fee & Over-Limit Fee Litig., 741 F.3d at 1028. Writing for the panel, Judge Nelson24×24. Judge Nelson was joined by Judges Reinhardt and M.D. Smith. noted that the “similarities and differences between liquidated [compensatory] damages and punitive damages . . . govern the outcome of this case,”25×25. Late Fee & Over-Limit Fee Litig., 741 F.3d at 1026. and enumerated the various features of punitive damages, in particular their goals of retribution and deterrence.26×26. See id. at 1027. Yet she decided the case not based on these features, but rather based on the fees’ origin: because they originate from private contracts, she concluded, they are distinct from jury-imposed punitive damages and, consequently, the Due Process Clause does not apply.27×27. See id. at 1026–27. Judge Nelson also noted the long history of courts rejecting contractual penalties on common law grounds rather than constitutional grounds. See id. at 1027.

Judge Reinhardt concurred in the judgment.28×28. Id. at 1028 (Reinhardt, J., concurring in the judgment). Judge Reinhardt was joined by Judge Nelson. He did so “reluctantly,”29×29. Id. conceding that the Supreme Court has not extended the reasoning of punitive damages cases to the contract context, but forcefully arguing that it should — at least for contracts of adhesion.30×30. See id. (arguing that the Court “would be well advised to apply the same rule to prevent disproportionate penalties from being imposed on consumers when they breach contracts of adhesion” as it does in the punitive tort damages context). Contracts of adhesion are “prepared by one party, to be signed by another party in a weaker position, usu[ally] a consumer, who adheres to the contract with little choice about the terms.” Black’s Law Dictionary 366 (9th ed. 2009). He focused on the ubiquity of adhesive contracts,31×31. Late Fee & Over-Limit Fee Litig., 741 F.3d at 1028 (Reinhardt, J., concurring in the judgment). the asymmetry of current doctrine in protecting corporate tort defendants but not consumers,32×32. See id. at 1030. and the importance of “[c]onstitutional evolution . . . to ensure that changes occur within the framework of fairness and equality.”33×33. Id. at 1029. He noted that contractual fees could constitute state action by dint of judicial enforcement.34×34. See id. at 1031 (“Substantive due process should . . . bar judicial enforcement of contractual penalty clauses when such clauses are so disproportionate to the damage caused by the breach that the damages would be impermissible as civil penalties in the tort context.”). In so arguing, Judge Reinhardt cited Shelley v. Kraemer, 334 U.S. 1 (1948), as precedent for finding state action in judicial enforcement of contracts. Late Fee & Over-Limit Fee Litig., 741 F.3d at 1031 (Reinhardt, J., concurring in the judgment). He acknowledged, but neither endorsed nor rejected, the plaintiffs’ theory of state action, which maintained that statutory and regulatory provisions abrogating the common law bar on penalty clauses constitute state action by allowing “imposition of . . . disproportionate penalties in violation of the Constitution.”35×35. Late Fee & Over-Limit Fee Litig., 741 F.3d at 1031 n.9 (Reinhardt, J., concurring in the judgment).

Extending the logic of punitive damages tort cases to the contract context requires an account of an issue only briefly discussed in the opinions: how the fees at issue implicate state action.36×36. State action analysis was not necessary to the Ninth Circuit’s decision, given that it found for the defendants on the merits. Cf. Aiken v. Ray, No. 9425331, 1996 WL 171439, at *2 (9th Cir. Apr. 11, 1996) (“Having affirmed the judgment below on the merits, we find it unnecessary to reach the state action issue.”). In addition to the theories advanced by Judge Reinhardt and the plaintiffs, public function theory could justify a finding of state action, given that punitive credit card fees allow private firms to exercise the power to punish — traditionally an exclusive state function. Though applying public function theory would require a substantial doctrinal change relative to current law, which has long rejected new public function claims, public function theory is nonetheless a normatively attractive means of finding state action. It would effectively capture the fact that the fees at issue are objectionable not only because of their amount or their having been authorized by the state, but also by virtue of the type of power that they allow card issuers to exercise.

The state action requirement stipulates that the Fourteenth Amendment applies only to government conduct37×37. See The Civil Rights Cases, 109 U.S. 3, 11 (1883) (“Individual invasion of individual rights is not the subject-matter of the [Fourteenth] [A]mendment.”). — namely statutes, regulations, actions by government employees, and a narrow range of private actions enabled by government.38×38. See generally Chemerinsky, supra note 5, § 6.4 (providing an overview of state action theory). In perhaps the most famous description of state action, Professor Charles Black dubbed it “a conceptual disaster area.” Charles L. Black, Jr., The Supreme Court, 1966 TermForeword: “State Action,” Equal Protection, and California’s Proposition 14, 81 Harv. L. Rev. 69, 95 (1967). Under the public function theory of state action, courts consider “the predominant character and purpose” of an activity in determining whether it should “be treated as a public institution subject to the command of the Fourteenth Amendment.”39×39. Evans v. Newton, 382 U.S. 296, 302 (1966). Public function cases of the mid-twentieth century treated some activities as “governmental in nature”40×40. E.g., id. at 299. and therefore state action even when performed by private parties.41×41. See, e.g., id. (operation of a public park); Terry v. Adams, 345 U.S. 461 (1953) (election administration); Marsh v. Alabama, 326 U.S. 501 (1946) (management of a central business district). Later decisions found that public function theory applied only when private entities exercised powers that are “traditionally the exclusive prerogative of the State.”42×42. Jackson v. Metro. Edison Co., 419 U.S. 345, 353 (1974) (emphasis added); see also Blum v. Yaretsky, 457 U.S. 991, 1005 (1982); Rendell-Baker v. Kohn, 457 U.S. 830, 842 (1982). While described as a “significant curtailment of the public function theory,”43×43. Daphne Barak-Erez, A State Action Doctrine for an Age of Privatization, 45 Syracuse L. Rev. 1169, 1176 (1995); see also Kathleen M. Sullivan & Gerald Gunther, Constitutional Law 683 (16th ed. 2007) (“The ‘public function’ analysis launched by Marsh was later cabined essentially to the facts of the company town and white primary cases.”). this exclusivity requirement was in fact the theory’s death knell — since imposing the exclusivity requirement, the Court has never classified a new function as exclusively public.44×44. Among functions found to be not exclusively governmental and therefore not per se state action are the provision of utilities, see Jackson, 419 U.S. 345; education, see Rendell-Baker, 457 U.S. 830; and nursing home services, see Blum, 457 U.S. 991.

Yet if the public function theory were still doctrinally robust, it could be applied to credit card fees by treating those fees as a form of punishment. Legal theorists have long viewed punishment as an exclusive government function.45×45. See, e.g., Hans Kelsen, General Theory of Law and State 21 (Anders Wedberg trans., 1949) (describing “the coercive act of the sanction” as “of exactly the same sort as the act which [the legal order] seeks to prevent in the relations of individuals”). Punitive damages, which can be understood as a form of punishment, have “throughout the legal history of England and the United States, been seated exclusively within the capacity of the state.”46×46. Karl E. Neudorfer, Defining Due Process Down: Punitive Awards and Mandatory Arbitration of Securities Disputes, 15 Ohio St. J. on Disp. Resol. 207, 243 (1999). A theory of punitive damages as the exclusive province of the state accords with at least some Supreme Court dicta, most notably the view that punitive damages “serve the same purpose as criminal penalties”47×47. State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 417 (2003). — the imposition of which is the paradigmatic exclusive government function. Moreover, courts could treat punishment as an exclusive public function even under a narrowed public function theory, given that punishment is — even more so than public goods provision — a traditional, exclusive, and constitutive state power.

The rules of contract law implicitly recognize the exclusivity of state control over punishment. In rejecting punitive contract provisions, the Restatement (Second) of Contracts notes that the “central objective behind the system of contract remedies is compensatory, not punitive.”48×48. Restatement (Second), supra note 12, § 356, cmt. a. As a result, contract law bars punitive damages, defined as those that are not “reasonable” relative to “the anticipated or actual loss caused by the breach.”49×49. Id. § 356. While scholars do not know for certain the reason for contract law’s longstanding bar on punitive damages,50×50. See William S. Dodge, The Case for Punitive Damages in Contracts, 48 Duke L.J. 629, 630 (1999) (“[T]he original reasons for the rule against punitive damages in contracts are obscure . . . .”). Contemporary scholars have invoked the theory of efficient breach to justify the bar on contractual punitive damages, but that theory does not explain the bar’s origin. See id. at 630–31. discomfort with a privately structured system of punishment provides one possible reason. Indeed, observers have noted that because “failure to perform . . . self-imposed responsibilities is only a breach of a private duty,” breach of contract “does not technically contravene social norms” and is therefore incompatible with punitive damages.51×51. Frank J. Cavico, Jr., Punitive Damages for Breach of Contract — A Principled Approach, 22 St. Mary’s L.J. 357, 368 (1990).

The Supreme Court has not expressly defined what renders sanctions punitive, but treating credit card late fees as punitive is in accord with at least some of the criteria it has set out. The Court has long understood punishment to be defined by the nature of the sanction, noting, in another context, that a sanction’s “character is not changed by the mode in which it is inflicted.”52×52. United States v. Chouteau, 102 U.S. 603, 611 (1880). Insofar as sanctions are punitive when they are disproportionate relative to actual damages,53×53. See, e.g., State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 425 (2003) (“Single-digit multipliers are more likely to comport with due process . . . than awards with ratios in range of 500 to 1 . . . .”); BMW of N. Am., Inc. v. Gore, 517 U.S. 559, 581 (1996) (“[A] comparison between the compensatory award and the punitive award is significant.”). credit card fees are almost certainly punitive.54×54. See Shiffrin, supra note 2, at 477–81. Yet if sanctions are punitive only when “imposed for purposes of retribution and deterrence,”55×55. E.g., State Farm, 538 U.S. at 416 (quoting Pac. Mut. Life Ins. Co. v. Haslip, 499 U.S. 1, 19 (1991)) (internal quotation mark omitted). the status of the fees at issue is more ambiguous. Card issuers impose fees with the intent of generating profits, not of punishing cardholders.56×56. Cf. Jessica Silver-Greenberg, Don’t Look Now, but Here Come the New, New Bank Fees, Wall St. J. (Nov. 6, 2010, 12:01 AM), http://on.wsj.com/9mUNsg [http://perma.cc/48FJ-UTYZ] (noting that some banks “rais[e] minimum payments on certain customers’ accounts in order to increase late penalties”). The primary effect of the fees on cardholders, however, remains retribution (because the fees convey a message of fault) and deterrence (because of fees’ effects on cardholder incentives).57×57. Despite courts’ pairing of retribution and deterrence, many observers “have worried that far from complementing each other, retribution and [deterrence] rely on principles that are actually inconsistent.” Marc Galanter & David Luban, Poetic Justice: Punitive Damages and Legal Pluralism, 42 Am. U. L. Rev. 1393, 1434–35 (1993). The Court has not addressed whether punishment, in the context of punitive fees, is defined more by intent or effect.58×58. Cf. Alice Ristroph, State Intentions and the Law of Punishment, 98 J. Crim. L. & Criminology 1353, 1372 (2008) (noting, in the criminal law context, that “[t]he doctrine allows, in theory, for a statute’s effects to take precedence over its intent: a law sufficiently ‘punitive in its effects’ will be subject to constitutional restrictions on punishment”). Yet the noncompensatory nature of the fees, coupled with their retributive and deterrent effects, is enough to render colorable the proposition that they are punitive — thereby opening the door to the application of public function theory.

Public function theory is the only approach to state action that captures one of the normatively objectionable features of the fees at issue: the power of card issuers to exercise state-like power over cardholders without due process constraints. Contractual penalty clauses are more similar to positive law imposed by the state than to terms negotiated by contracting parties: cardholders do not author contract terms, are largely unaware of those terms,59×59. See, e.g., Margaret Jane Radin, Boilerplate: The Fine Print, Vanishing Rights, and the Rule of Law 22 (2013). and consent to them only in the most attenuated fashion.60×60. See id. at 30 (arguing that, in attempts to justify boilerplate terms under a consent theory, “consent is degraded to assent, then to fictional or constructive or hypothetical assent, and then further to mere notice . . . until finally we are left with only a fictional or constructive notice”). Consent-based arguments were nonetheless central to the circuit court decision, see Late Fee & Over-Limit Fee Litig., 741 F.3d at 1026–27; defendants’ briefing, see Brief for Defendants-Appellees at 29–37, Late Fee & Over-Limit Fee Litig., 741 F.3d 1022 (No. 08-15218); and the district court decision, see In re Late Fee & Over-Limit Fee Litig., 528 F. Supp. 2d 953, 959 (N.D. Cal. 2007). In each of these respects, adhesive contracts diverge from the classical contract model of bargained-for exchange. The power exercised by firms authoring adhesive contracts, particularly when those contracts grant firms the power to punish, is thus analogous — normatively, even if not under current doctrine — to state power.61×61. Additionally, while the fees at issue are authorized by positive law, in other contexts adhesive contracts effectively displace relevant law. See Radin, supra note 59, at 16 (describing the “democratic degradation” of “delet[ing] rights that are granted through democratic processes, substituting for them the system that the firm wishes to impose”). The most notable examples of this displacement are the Court’s decisions on mandatory arbitration clauses. See Am. Express Co. v. Italian Colors Rest., 133 S. Ct. 2304 (2013); AT&T Mobility LLC v. Concepcion, 131 S. Ct. 1740 (2011).

The firm-state analogy is challenged by the fact that, at least in theory, punitive fee provisions are avoidable by simply declining to obtain a credit card, or by declining to obtain one with certain terms. An application of public function theory to private contracts could thus find state action only when the relevant contract terms are de facto unavoidable.62×62. Unavoidability is the mirror image of the classic contract scenario, in which “[o]ppressive bargains can be avoided by careful shopping around.” Friedrich Kessler, Contracts of AdhesionSome Thoughts About Freedom of Contract, 43 Colum. L. Rev. 629, 630 (1943). Credit card penalty fee provisions may meet that high threshold — credit cards, unlike most consumer goods, are a “practical necessit[y] of modern life,”63×63. Late Fee & Over-Limit Fee Litig., 741 F.3d at 1028 (Reinhardt, J., concurring in the judgment); see also Shaila Dewan, Buying Street Cred, N.Y. Times, Feb. 16, 2014, at MM14 (documenting numerous ways in which “[l]ife without credit is . . . potentially ruinous”). and the vast majority of credit card contracts contain comparable penalty fee provisions64×64. See Complaint at 21, Late Fee & Over-Limit Fee Litig., 528 F. Supp. 2d 953 (No. 4:07-cv-00634) (documenting that, when litigation commenced, defendant banks, which controlled over seventy percent of the market, imposed nearly identical late fees). — but additional facts about the market structure would be necessary to establish unavoidability.

Requiring that a particular type of fee be unavoidable before applying the public function theory would have several benefits. First, and most significantly, it would severely limit the effects of a new rule, thereby preserving private ordering in nearly all contract law cases, even those involving contracts of adhesion.65×65. Narrowing of private ordering is a primary fear among skeptics of expanding state action doctrine. See, e.g., Mark D. Rosen, Exporting the Constitution, 53 Emory L.J. 171, 191 (2004) (“Shelley’s holding was troubling . . . because, under its reasoning, every private contract that a party wishes to judicially enforce triggers state action . . . .”). Second, a rule turning on unavoidability could draw on jurisprudence that courts have developed in several other areas of law.66×66. Courts have characterized as unconscionable contracts entered into under “an absence of meaningful choice.” Williams v. Walker-Thomas Furniture Co., 350 F.2d 445, 449 (D.C. Cir. 1965). In constitutional law, the Court has similarly evaluated contextual factors, including economic factors, in determining when ostensible choices are in fact coercive. See, e.g., Nat’l Fed’n of Indep. Bus. v. Sebelius, 132 S. Ct. 2566, 2601–08 (2012) (opinion of Roberts, C.J.). Finally, requiring a determination of unavoidability, while not directly rooted in precedent, would follow from the Court’s invitation, albeit in an earlier era, to make fact-specific determinations in state action cases.67×67. See Burton v. Wilmington Parking Auth., 365 U.S. 715, 722 (1961) (“Only by sifting facts and weighing circumstances can the nonobvious involvement of the State in private conduct be attributed its true significance.”).

Though novel and counter to strong doctrinal trends, relying on the public function theory to impose restraints on firms that wield state-like power over consumers would be consistent with one of the Court’s longstanding commitments: interpreting the Fourteenth Amendment to protect those who are structurally disadvantaged by the political process.68×68. See, e.g., United States v. Carolene Prods. Co., 304 U.S. 144, 153 n.4 (1938) (foreshadowing courts’ “searching judicial inquiry” of laws that “tend[] seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities”). One scholar has noted that this theory “unifies some of the greatest successes in the Court’s history.” David A. Strauss, Lecture, Is Carolene Products Obsolete?, 2010 U. Ill. L. Rev. 1251, 1251. In the credit card context, the government abrogated the generally applicable bar on punitive contractual damages in response to interest group pressures.69×69. See, e.g., Vincent D. Rougeau, Rediscovering Usury: An Argument for Legal Controls on Credit Card Interest Rates, 67 U. Colo. L. Rev. 1, 10 (1996) (documenting card issuers “pressur[ing] their home state legislators into liberalizing the usury statutes”); Patrick McGeehan et al., Soaring Interest Is Compounding Credit Card Woes for Millions, N.Y. Times, Nov. 21, 2004, at N1 (documenting industry litigation and lobbying over interest rates). In its jury-award decisions, the Court took account of power dynamics in justifying its application of the Due Process Clause. See, e.g., State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 417 (2003) (fearing that jury decisions will reflect a bias against businesses, especially ones without local presences). Applying the Due Process Clause to credit card fees would thus serve as a firewall to protect consumers from firms that, because of that successful advocacy, now exercise state-like power in their ability to introduce punitive fee provisions into adhesive contracts. Public function analysis highlights that, though far from exercising the full complement of sovereign powers, private firms can at times look more like state actors than is commonly realized.